Join Kalray’s CTO, Benoit Dupont de Dinechin, as he delivers a presentation: “Supporting Standard CNN Networks on Manycore Processors”
Synopsis: KaNN is a domain-specific code generator for CNN inference that targets manycore processors featuring software programmable cores, local and DDR memories. Input is a standard CNN description such as Berkeley Caffe prototxt file, augmented with the parameters obtained during training. The KaNN generated code is optimized for low-latency parallel execution and effective DDR bandwidth exploitation. This result is obtained through the exploration and the selection of one of the many compute graph that represent the CNN forward computation, and the evaluation of this graph across the DDR and local memories according to a macro-pipelining scheme. The KaNN compute graphs have nodes that correspond either to 3D tiles of the images that represent the results of layer processing, or to operators on these tiles including 1×1 convolution kernels, pointwise accumulation and concatenation. The generated code is a collection of C kernels with asynchronous DMA transfers between the local and DDR memories. Kalray will be presenting KaNN and how it has been built to execute efficient CNN.
For more information click here.